Receptor recognition is the first step in viral infection and plays an essential role in target-cell selection in the infected host. Many viruses use cell-adhesion molecules or cell-surface carbohydrates as receptors. However, general rules governing receptor recognition at an atomic level have not been established, and contributions of multiple receptors to viral attachment and cell entry are poorly understood. The proposed research uses reovirus, a highly tractable experimental model that shows promise for oncolytic and vaccine applications, to define the structural basis of virus-receptor interactions at atomic resolution. Following primary infection in the murine intestine, reovirus disseminates to the central nervous system (CNS), where it exhibits serotype-specific differences in tropism and pathogenesis attributable to viral attachment protein 1. The 1 protein is a filamentous trimer consisting of an N-terminal tail and a C-terminal head. The 1 tail of strain T3D reovirus binds sialic acid (SA), and the 1 head of all three reovirus serotypes binds immunoglobulin superfamily receptor junctional adhesion molecule-A (JAM-A). Three integrated specific aims are proposed to define the structural and functional basis of 1 interactions with its receptors. In Specific Aim 1, structures of the three serotypes of 1 in complex with JAM-A will be determined using X-ray crystallography. Residues in each serotype required for JAM-A binding will be identified by structure- guided mutagenesis of intact virus using a newly developed plasmid-based reverse genetics system. The role of JAM-A binding in reovirus tropism in the murine CNS will be defined using mutants altered in JAM-A utilization and primary cultures of ependymal cells and neurons. In Specific Aim 2, the structure of T3D 1 in complex with SA will be determined using X-ray crystallography. Carbohydrate ligands of the three serotypes will be identified using glycan array screening and functional assays. Minimum sequence units required for carbohydrate binding in strains T1L and T3D 1 will be defined using chimeric viruses and assays of viral binding and infectivity. In Specific Aim 3, functional relationships between the 1 receptor-binding domains in reovirus attachment and cell entry will be elucidated using mutant viruses with alterations in 1 flexibility and length. Adhesive properties of the SA-binding region and its interaction with JAM-A binding will be determined by engineering additional SA-binding sites into the 1 tail. The 1 head will be replaced with the adenovirus fiber knob to define the function of receptor specificity in reovirus binding, internalization, and disassembly. These studies will enhance a basic understanding of mechanisms by which pathogenic viruses engage cellular receptors and accelerate the rational design of viral vectors for therapeutic purposes.